Simultaneous observations of families and accompanied air showers with emulsion chambers and the air shower array of electronic equipment at Mt. Chacaltaya ͑5200 m, 540 g/cm 2 ͒ reveal that families bear the data of nuclear interactions generated deep in the atmosphere. 47 outstanding families with ⌺E ␥ у10 TeV are correlated with the accompanied air showers of the size 10 5 -10 8 . A scatter plot of the average family energy versus the size of the relative air shower requires further energy fractionizing process͑es͒ in the propagation of high energy cosmic rays in the atmosphere, such as a larger dissipative mechanism in nuclear interaction, heavier chemical composition of the primary cosmic rays, etc. We reach the conclusion that nuclear interaction changes its features in the energy region E 0 Ͼ10 16 eV, because the heavier composition, proposed so far, is not sufficient for the required dissipative process. A comparison with the data from the HADRON experiment at a similar altitude with a similar technique shows that no larger deviations are present between both experiments.

Results of Monte Carlo simulations for a combined emulsion chamber and extensive air-shower array experiment are shown. Two totally distinct models are assumed for nuclear interactions, one based on the UA5 Monte Carlo algorithm (model A) and the other including exotic processes of Centauro and MiniCentauro (model B). Effects of different assumptions on the chemical composition of primary particles are also considered. Longitudinal and also lateral structures of gamma-families triggered by air-shower size, Ne, are studied and are compared with the data of the HADRON experiment at Tien-Shan and of the SYS experiment at Chacaltaya. The experimental data are found to disagree with the results of model A, even when we introduce the heavy-dominant assumption for primary cosmic rays. On the other hand, model B explains the general tendency of the experimental characteristics. It suggests that the nuclear interaction is changed drastically in the primary energy range over 1016 eV.

The cosmic-ray exotic event ''Centauro-I'' is reexamined. Kopenkin et al. [Phys. Rev. D 68, 052007 (2003).] have recently suggested that the original correspondence of shower clusters in the upper and the lower chambers [C. M. G. Lattes, Y. Fujimoto, and S. Hasegawa, Phys. Rep. 65, 151 (1980).] is not correct, and we confirm this suggestion. By taking into account the relative position of the upper and the lower chambers exactly, we find that the event has no upper part corresponding to the observed shower cluster in the lower chamber. Our analysis shows that showers in the shower cluster in the lower chamber are likely to have originated in a bundle of target interactions (C-jets) and that they are not a simple atmospheric family as was claimed by Kopenkin et al. The event shows peculiar characteristics quite different from commonly observed cosmic-ray events.

No abstract

No abstract